Dozy DMV Wants More Dollars

In the state capital of Sacramento, a man named Tyré Nichols went online, secured all his paperwork, and arrived promptly at the Department of Motor Vehicles expecting to be in and out in two hours. Despite his preparation and punctuality, Nichols wound up waiting six hours, “in miserable 98-degree heat.” As it happens, this is a typical experience. Over the last year, DMV wait times have increased 60 percent, delaying Californians by up to seven hours. While Californians wait in line, DMV employees sleep on the job.

As we noted, according to California’s state auditor, “A key data operator at the Department of Motor Vehicles failed to perform her essential duties over a period of nearly four years because she slept at her desk for extended periods of time during work hours. From February 2014 through December 2017, the employee misused more than 2,200 hours of work time as a result of sleeping on the job, costing the state more than $40,000.” The unnamed employee still works at the DMV, her job protected by the powerful Service Employees International Union, which parades outside the capitol chanting “This is our house!” Other DMV employees have been using Californians’ personal information to commit bank fraud and identity theft, and DMV bosses have taken bribes to alter records and hand out commercial driving licenses. 

These lapses have prompted legislators to demand an audit, but DMV boss Jean Shiomoto is opposed. “Any audit you have to pull a lot of documents,” she told reporters, “It would strain our resources.” Shiomoto is requesting an additional $26 million to hire 400 new employees, but she announced no new policy of dismissing those employees who sleep on the job for extended periods over a span of four years. According to insiders, this sort of sloth is common. Until the DMV starts showing deadbeats the door, a ballpark figure for additional funds should be zero. Legislators should conduct the audit whether or not Shiomoto wants it, and if she resists they should show her the door without delay. 

***

K. Lloyd Billingsley is a Policy Fellow at the Independent Institute and a columnist at The Daily Caller.

copy top traders

Stopping the Revolving Door in DC

You may be familiar with CEPR’s economic research. But did you know that there is a program at CEPR that works to ensure that political appointees are focused on serving the public interest, rather than personal professional advancement?

Practically alone in Washington DC, the Revolving Door Project (RDP)—based at CEPR—applies pressure on both Democrats and Republicans to administer the executive branch on behalf of the common interest rather than personal interest. From Wall Street insiders seeking jobs in a potential Hillary Clinton Administration to Steven Mnuchin, Mick Mulvaney, Wilbur Ross, and the rest of President Trump’s ethically compromised team, the Revolving Door Project has a proven track record of identifying and scrutinizing the selfish.

“The deepest rules of our rigged economy are usually written by people who hold key but obscure jobs within the executive branch. From the Treasury Department and the Federal Reserve to the CFPB, FTC, and SEC, the Revolving Door Project highlights corrupt personnel while pushing for appointees committed to fighting for the public interest rather than cashing in.”

— Jeff Hauser, Director of the Revolving Door Project

With the Trump Administration continuing to staff the executive branch with the greediest among us, we really need your support for this crucial program. Won’t you consider making a donation to CEPR to help fund this work? We rely on the generosity of individuals like you to fund our research, analysis, media work, outreach – everything we do.

Thanks in advance for your support!

copy top traders

Cryptocurrency Malware Infects Over 200,000 Mikrotik Routers

Cryptocurrency Malware Infects Over 200,000 Mikrotik Routers

A cryptojacking campaign has affected over 200,000 routers made by Mikrotik, the Latvian networking company. 


A Months-Old Vulnerability Exploited

Security researchers recently mapped a series of cryptomining attacks, which initially attacked a large number of users in Brazil to create a growing mining botnet by infecting compromised devices with malware.

According to reports, the devices targetted for the attack were Mikrotik routers which had an outdated software patch.

In April 2018, the company patched a remote access vulnerability which allowed attackers to remotely gain unauthenticated administrative access to the Mikrotik routers.

Some security researchers who reverse engineered Mikrotik’s patch then published a proof-of-concept exploit explaining how to use the recovered vulnerability to access Mikrotik devices.

Some security researchers who reverse engineered Mikrotik's patch then published a proof-of-concept exploit explaining how to use the recovered vulnerability to access Mikrotik devices.

This information was used to infect the routers with code that loads the CoinHive browser-based cryptomining software.

This happens whenever users accessing the internet through the routers encounter an HTTP error and they are browsing via the Mikrotik proxy.

Coinhive’s Javascript is injected into web pages accessed by users on a compromised router. The users then mine Monero for the attackers without any knowledge.

A Cryptojacking Threat That’s a Global Threat

There have been at least three cryptojacking attacks from this vulnerability that have been noted by researchers so far. The first was recorded in Brazil and it reportedly affected more than 183,700 MikroTik routers.

Two other attacks that affected 16,000 and 25,000 MikroTik routers respectively mainly in Moldova were also recorded by another security researcher.

This indicates that this campaign that isn’t limited to one specific geographic region, which has worried analysts and researchers amid an overall growing trend.

Cryptojacking cases have exploded over the past couple of years and are emerging as one of the primary cybersecurity threats around the world, with cases on the rise even for traditionally safer operating security systems like Linux.

As is always the case around cybersecurity, users are being urged to be vigilant especially when accessing public networks. Analysts in the cybersecurity space have also been very clear; If you have a Mikrotik device apply a patch immediately and update any passwords.

Have you been a victim of the MikroTik router attack or any other cryptocurrency mining hack? Share your experiences in the comments below.


Images courtesy of Mikrotik.com, Shutterstock

The post Cryptocurrency Malware Infects Over 200,000 Mikrotik Routers appeared first on Bitcoinist.com.

binary options

Decred Mining Explained

Forex stock trading

Decred Mining Explained

Decred is a decentralized cryptocurrency that is well-known for its blockchain integrated community-based governance features. Decred achieves this by use of its hybrid proof of work (PoW) and proof of stake (PoS) consensus mechanism. This produces a system in which stakeholders are able to control the development of the Decred blockchain, as Decred’s hybrid proof of work and proof of stake consensus mechanism allows for the ordered transition from one set of consensus rules to the next. The Decred mining and staking system also helps to ensure that the progression of the project is not controlled by powerful self-interested groups that can initiate changes to the underlying blockchain without the input of the community.

Decred Mining – Proof of Work

Decred mining first concerns the proof of work element, which involves individuals investing their computational resources in order to form blocks for the blockchain and process transactions on the network. Each time a block is mined, 30 Decred coins (DCR) are released, these coins are then further subdivided in the following way:

  • 60% PoW Miners
  • 30% PoS Voters
  • 10% Decred Development Subsidy

Decred Mining – Proof of Stake

The second element to Decred mining is proof of stake, which is designed to serve the following:

  • Allow Decred stakeholders to vote in favour or against proposed changes to the blockchain. If stakeholders vote to support a change, the Decred blockchain will hard fork and the new feature will become live on the network.
  • Serve as a mechanism by which stakeholders can act as a counterbalance to malicious proof of work miners. Individuals involved in staking can vote a block invalid even if it complies with Decred’s network consensus rules.
  • Reward individuals that help to secure the network through the staking process by awarding them 30% of the total block reward.

In order to participate in the proof of stake system, individuals must lock up some DCR in exchange for a ticket, with each ticket representing a single vote in the Decred system. These tickets serve as a mechanism by which holders can participate in the governance of Decred. Once a user purchases a ticket, it enters a lottery pool. With every block that is discovered by a proof of work miner, 5 randomly chosen tickets must be included in the block (Miners face penalization in the form of a reward deduction if fewer than 5 votes are included in a mined block). These 5 randomly chosen tickets must now vote on whether or not to approve the previously mined block. If at least 3 votes approve the previously mined block, then the newly mined block is added to the chain, and both the miners and stakers are rewarded. However, if the previously mined block is not approved, then miners are not rewarded, but stakers are. The rationale behind this is to incentivize miners to mine in accordance with the wishes of stakers.

Decred mining, more specifically the proof of stake element, rewards every ticket that is chosen to vote in a block. Each randomly selected ticket is given an average vote time of 28 days. If a ticket expires before casting a vote, then the original ticket price is returned to the purchaser.

The proof of stake system serves as a check on problems that could result from proof of work only systems:

  • Miners cannot decide to change the rules of the network (e.g. a 51% attack) because stakers must validate the blocks that are discovered by miners.
  • In the event of a Decred hard fork, the old chain will quickly fall into disuse due to blocks not being validated by stakers on the chain.

Decred Tickets and Ticket Price

Whenever a user purchases a ticket on the Decred network, they must pay an internal ticket price and a ticket fee using DCR. Purchased tickets are then transferred to a temporary mempool. Tickets in the mempool must wait until they are chosen to be included into a block, with a maximum of 20 tickets capable of being mined into a block. As there is a limit to the number of tickets that can be mined into a block, competition to get tickets mined will result. A user can decide to pay a higher ticket fee to miners on the network to get their ticket mined more quickly. Once a ticket is included into a block, it is moved to the ‘immature’ ticket pool. After a waiting period of 256 blocks (approximately 20 hours), the ticket will mature and go into the live ticket pool, also known as the lottery pool. Once in this pool, the ticket is eligible to be chosen to validate blocks on the Decred blockchain.

The price of tickets on the Decred network is determined by an internal algorithm that is configured to keep the ticket pool, which is the total number of tickets in the proof of stake system, around a target size of 40,960 tickets. The price of a ticket will fluctuate in accordance with the demand for a ticket, and the supply of tickets that currently exist in the pool.

The algorithm adjusts the price of a ticket every 144 blocks, this is known on the Decred network as the ‘buying window’. With each block capable of housing 20 tickets, a buying window can result in a maximum of 2,880 tickets being purchased. The ticket price for a purchased ticket is always refunded regardless of if the ticket casts a vote or expires.

Decred Mainnet Voting

There is a two-stage process for voting to implement consensus changes that would result in a hard fork of the Decred blockchain. However, it is first important to note that two different block intervals exist for the voting process.

  • The Stake Version Interval (SVI) of 2016 blocks (approximately 1 week)
  • The Rule Change Interval (RCI) of 8064 blocks (approximately 4 weeks)

The first stage in the voting process is that the upgrade threshold on the network must be met. When the code that will hard fork the blockchain is released, a majority of nodes participating in the proof of work and proof of stake systems must first upgrade before voting can begin.

For nodes on the proof of work system, at least 95% of the 1000 most recently produced blocks must have the latest block version. For nodes operating on the proof of stake system, at least 75% of the votes cast within a single SVI must possess the latest vote version. Once these respective upgrade thresholds are met, voting can begin on the first block of the next RCI.

The second stage in the voting process is to vote:

  • If more than 90% of all votes submitted within the RCI are Abstain votes, then the agenda vote will remain active for the next RCI.
  • If all non-abstaining votes submitted within the RCI fail to meet the 75% Yes or No majority threshold, then the agenda vote will again remain active for the next RCI.
  • If 75% of all non-abstaining votes within the RCI vote in favour of the agenda (Yes vote), then the agenda will be locked in and the consensus change will activate 8064 blocks after the vote passed.
  • If 75% of all non-abstaining votes within the RCI are against the agenda (No vote), then the it will fail and the consensus change will never activate.
  • If an agenda reaches its expiration before reaching a 75% majority vote, then it will expire and the consensus change will never activate.
Source: Decred.org


 

The post Decred Mining Explained appeared first on Mycryptopedia.

What you should know before working in Cybersecurity in the Military.

Forex stock trading

Photo by rawpixel on Unsplash

Heard you are working in the weapon industry now ? How is sleeping at night going for you?

This sentence was addressed to me a few years ago by a friend’s dad while his daughter was experimenting every possible drugs in her room 20 meters away from him, slowly trying to kill herself. As astonished as I was that this dad was more concerned about my ethics than his daughter life, this is actually the first time I seriously considered ethics in development as a part of my duty as a developer.

Just after graduating in 2012, I had started working for the service branch of a French multinational that happens to develop weapons (Cyber, Electronic and conventional weapons). Working in the Network and Systems security branch I started as a Software Engineer, designing and building applications mainly for Banks and Insurances and never had faced any ethical dilemma other than the good old: “Should I tell the manager that this user has been watching adult content with the company’s device”. Good. I can handle that. I intended to keep it that way. After all, the only reason why I had chosen this company as a first job was because the pay was really good and that everybody wanted to work there. Anyway I had other plans for the future and just wanted to learn as much as possible and be financially comfortable to start something on my own later.

This company’s main client was the state military (or defense, depending how you see it) and although my job was like any other Software engineer at the beginning I must say I took advantage of it a number of times during random social encounters

I work for the military as an engineer, most of my projects are classified so if I tell you about it I will have to kill you after.

That worked great as a conversation starter! Much better than the “I am a Software Engineer” which at that time in my city was usually welcomed with at best a polite “sigh” at worst an eye roll and a “Eww, so you are a nerd”.

But internally to cope with the little guilt I was feeling I would tell myself.

It’s ok you just develop Web Applications, it’s never gonna kill anybody

And it was true, as good as they were my forms and buttons could not be possibly used in any lethal way .

Working for the defense

Eager to learn, and working day and night I was well considered by my pairs and I climbed the company’s ladder. Gradually I was offered much more interesting missions that were closer to one of their core business : Cybersecurity.

Maryland Guard cyber warfare operators from the 175th Wing’s Cyber Operations Group

Working for Cybersecurity was way cooler than everything I had done before. I was working in a research lab, we were free to innovate, to spend as much time as we needed trying out the latest technologies and building prototypes. Eventually we developed an internal framework combining cutting edge web technologies that was used to showcase the latest Cyber-defense innovations during European defense shows.

These prototypes would range from a simple tool to visualize in real time the vulnerabilities in a complex network to a complete air defense control app where you could see in real times the radars coverage and visualize the trust you have in these data from a Cyber point of view.

Again I was asked if I was developing weapons

Nah, I only develop the visualization tools for threat detection and these are only prototypes anyway.

I was promoted as an expert in web software technologies and I never went further since I left the company shortly after to start my own business . But I am sure that as I would have progressed in the company I would have ended up on more sensitive projects or actually implementing the real systems that I had been prototyping. And here is my question: where does your responsibility as a developer starts ?

Military despite yourself

Photo by Hello I’m Nik on Unsplash

Few people know but latest missiles can be compared to supercars. They cost millions each (thank god), are auto-guided, on-board a computer, the latest high definition cameras, dozens of sensors and use cutting-edge technologies. Most of the time these sensors and technologies were not primarily designed for military use and the people that have conceived them are not even aware of how it is being used. Designing and assembling these missiles probably only requires few people, but in reality I would not be surprised if more than 1000 people have worked on the embedded technologies that will allow target detection, missile launch, guidance and impact optimization . Actually if you are contributing to a widely used open source software or library your lines of code are probably already on there. In my example, maybe one library that I’ve prototyped will be re-used in an air monitoring system that will allow threat detection and will automatically trigger a counter-attack. Of course I was working in the military so even if I never thought about it this way before, I agreed to my code being re-used for any purpose when I signed the contract but that is not the case for most of us in the open source community.

Most popular open sources licenses and NMPL license concept : Non Military Public License

My way to cope with that after leaving the company was to ask myself “Could what I’ve built be used for doing any harm?” but I quickly realized that pretty much everything you do can be used in a bad way (You can kill someone with a Pretzel after all), and the bigger impact your contribution has the more dangerous it is. So I just limit myself to the direct impacts that my work could have for lack of a better solution.

So I was thinking, instead, could we start a new license that will allow opening sources and forbid re-use for the military? “Non Military Public License” or something like that. And can it realistically be controlled ?


What you should know before working in Cybersecurity in the Military. was originally published in Hacker Noon on Medium, where people are continuing the conversation by highlighting and responding to this story.

Framework to build an MVP (Minimum Viable Product)

Forex stock trading

Everyone has an idea, but getting started is the most confusing of all, because you are trying to optimise too many variables around your idea. Here I will share with you a simple framework to get started with your product on a shoestring budget and a very quick turnaround time. I will focus on giving you a framework for digital products but you can use it for any product idea that you may have.

This is what it looks like : First and foremost, you decide the bare minimum features that your product should have which makes the product usable for users and at the same time helps in shipping the product to the market at the earliest. The idea behind this is to collect user feedback on the product without wasting any time and money. This version of the product is called MVP — Minimum Viable Product.

Lets talk about what an MVP is.

Eric Ries defines MVP as “ that version of a new product which allows a team to collect the maximum amount of validated learning about customers with the least effort.” MVP as a concept has been popularised by lean startup methodology. You can read more about it in this book by Eric Ries — The Lean Startup.

Let’s take Whatsapp as our example, in order to understand MVP better. Whatsapp is one of the most popular messaging apps today with a user base of over one and a half billion. In 2009 It was launched as an app which would publish current notification of a person to entire network, that was the only feature available in the app, then. This was to make use of newly launched push notifications by apple.

In the next version of Whatsapp that was released 3 months later, messaging was introduced in the app, this made a huge impact. Make a note, this was just text messaging, sharing of images and videos were enabled later on. Voice Call was added in the app as late as 2014.

Whatsapp is a perfect example of lean startup, the engineering team before getting acquired by Facebook was as small as 32 engineers. They managed and served millions of users with this time without any significant outage or downtime.

Every product that you use today will have a similar story, Facebook, which was launched in 2004, did not have a mobile app as recent as 2009. The Facebook that you use today was not built and launched in one go!

Make something people want

The goal of building a product is to build something that users will use.
“Make something people want, that’s the fundamental problem. If you die, it’s probably because you didn’t make something people wanted” says Paul Graham, co-founder of Y-Combinator.

Building a no-frills version of your product not just helps in fast go-to-market, it also makes marketing of the product easier. It helps in getting user feedback quicker. You can quickly iterate over the product based on the feedback you receive and launch in no time.

This, on one hand, saves time while on the other it helps you save cost, faster turnaround ensures that you do not burn your pockets beyond repair. The dollars saved here can be utilised in better marketing of the product. It’s always wise to build quick, fail fast and iterate.

Validate your idea before you build an MVP

You should build an MVP when your idea is relatively new and you want to test waters before going full in. When the idea has not been validated by actual users or customers, it is better to start small and keep building on it iteratively.

But in cases where you already have product-market fit (you have enough number of user transaction on your product to prove that the product is needed by the end users.), you do not need to build an MVP.

The decision to build an MVP or go with a full-blown product first up depends on the stage at which you are with your idea, as well. For a matured product idea, you obviously do not necessarily need to take the MVP route.

Metrics to decide what goes in the MVP

Now, I will give you simple metrics based on which you can decide what needs to be built and taken care of while building this version of your product. Start with making a list of all the features that you will want to have in your product. Categorise these features into two buckets, one for “product does not exist without this” and another “good addition to the first list”. More than anyone else, this will give you an understanding of what core value your product is offering and here on every decision that you take becomes easier. The MVP of the product will have only those features which are a must-have, without these features, there is no existence of the product itself.

In initial days, you should target one specific set of customers, who would be willing to use the product and pay for it, its this segment of users that you are building the product for. When Facebook was launched in 2004, it targeted only students from Harvard University. Then they moved on to other universities before opening up for everyone else. This helped gain valuable insights and helped serve the initial smaller set of target audience really well.

What goes in your MVP also depends on the market you are targeting, I am not talking about geo-location as such, its more about the maturity of the market. In an already served market, what you build will be assessed against what your competitors have already built. While in an untapped market, you will not be judged that critically.

Summary

A common misconception around MVP is that it means an incomplete product. MVP means lesser features where your user flows are complete but you are lesser focused on.

In cases where you need to pivot based on user feedback, the product marketing costs increases and time gets prolonged as you continually have to change the marketing pitch. When the product that you have built has a definite value and has seen a significant number of transactions by users, it’s wasteful to build an MVP.

MVP provides you with the liberty to focus on the most critical features first. While you work on subsequent features, your idea is further validated by users. Hence, the final product would not only be superlative in terms of user experience but would also be tailor-made for the target audience as it would accommodate user feedback.

From our experiences with founders and entrepreneurs, it’s a good idea to start with MVP and pivot as per the customer feedback till you reach a product-market fit.

Originally published at sugoilabs.com on June 20, 2018.


Framework to build an MVP (Minimum Viable Product) was originally published in Hacker Noon on Medium, where people are continuing the conversation by highlighting and responding to this story.

The Bitcoin ETF: breaking down the CBOE ETF proposal

Forex stock trading

A bitcoin Exchange-Traded Fund (ETF) would be massively bullish for bitcoin’s price. It would further legitimize bitcoin and would allow several asset managers, currently precluded, to participate in the asset class.

TL;DR

The CBOE bitcoin ETF proposal is vastly superior to prior ETF proposals, and addresses most of the concerns the SEC has expressed when rejecting prior ETF applications. Yet the SEC’s key concerns remain: underlying bitcoin markets are not demonstrably resistant to manipulation, and are mostly unregulated. An ETF needs to require rules to prevent manipulative practices. For this, an exchange must have sharing agreements with significant markets for the underlying commodity, and those markets must be regulated.

However, one of the SEC commissioners has already expressed support towards a bitcoin ETF. My assessment is that it is feasible the SEC will decide the CBOE ETF proposal does satisfy requirements — even if it doesn’t. It should be a long process. I do not expect the SEC to reach a final decision before the final deadline, which is Feb/27/2019.

Table of Contents

This article is an in-depth analysis of the CBOE bitcoin ETF proposal. It is structured as follows:

  • Why an ETF is a game changer.
  • Estimating the price impact of a bitcoin ETF.
  • Hold your horses.
  • The “CBOE bitcoin ETF”.
  • SEC concerns regarding bitcoin ETFs.
  • Advantages & shortcomings of CBOE proposal.
  • Comparison with CME & CBOE bitcoin futures.
  • An SEC commissioner supported the Winklevoss ETF.
  • Timing the SEC announcement.
  • Trading the SEC announcement.
  • Conclusion.

Why an ETF is a game changer

ETFs are exchange-traded funds whose shares track the price of an underlying asset (e.g. commodities such as gold) or underlying basket of assets (e.g. components of the S&P 500 index). ETFs are popular investment vehicles for a myriad of reasons. One of the key advantages of commodity backed ETFs is allowing investors to not worry about custody of the underlying assets (bitcoin here represents a commodity), as the ETF takes care of custody.

Back to bitcoin, institutional investors in particular are generally uninterested in buying bitcoin precisely because of the associated custody issues.

As an alternative to physical bitcoin, investors could buy bitcoin futures, financial derivatives based on bitcoin which require no custodianship. Yet futures need to be rolled over every time a futures contract expires, which is cumbersome, costly, and results in a taxable event. Furthermore, smaller asset managers such as corporate pension plans are precluded by mandate to buy futures directly — all futures, including bitcoin futures.

A bitcoin ETF would thus create a new investable asset class for many institutional market participants. A game changer! This game changer needs to first be approved by the U.S. Securities and Exchange Commission (SEC), the US regulator overseeing securities.

Estimating the price impact of a bitcoin ETF

How big of a price impact could a bitcoin ETF have? I’ll use the case of gold to come up with an estimate for bitcoin.

The total gold market size is estimated at 188,268 tonnes (World Gold Council, Dec/2017). The gold market is composed of five sectors: Jewelry, Official Sector (i.e. central banks), Bars and Coins, ETFs and similar, and Other Fabrication and Unaccounted. For simplification purposes, one could argue gold ETF and Bars and Coins are the two strictly speculative categories. Given that the gold ETFs market size (2,158 tonnes, or $75 billion at current gold prices) represents 5.6% of the Bars and Coins market size (38,444 tonnes), one could then theorize that the creation of a bitcoin ETF may attract an influx of $6,7 billion into the bitcoin markets (that is, 5.6% of the bitcoin current market cap of $120 billion).

Then comes the question of how much of an impact would that influx have on bitcoin’s price. The relationship is not 1 to 1. Cheesman and Burniske, in their pieceCryptoassets: Flow & Reflexivity”, called that relationship the fiat multiplier, and estimated a multipler in the 2x–25x range.

Reflexivity is most commonly thought of as an internal feedback loop, where investor perception becomes a self-reinforcing reality. As a result, asset prices can be fed solely by the signals of other investors, or variables endogenous to the process of investing, as opposed to exogenous variables that are more representative of an asset’s fundamentals.

In other words, an influx of $6,7 billion would have a considerably larger impact on price because of investor’s feedback loop and associated price momentum. By applying a multiplier of 13.5x (the median of such range), the 5.6% fiat influx would result in a 76% price increase.

Another approach is looking at the price of gold after the inception of the first gold-backed ETF in the United States, $GLD, on Nov/18/04. Gold skyrocketed soon after, clocking a 318% increase in approximately seven years. However, it is important to bear in mind there was a strong monetary policy component propelling gold prices upwards at the time.

(For more information on how gold ETFs transformed the gold market, refer to this excellent article — drawing similarities with bitcoin should be straightforward.)

Hold your horses ✋

Bullish factors aside, several asset managers are precluded from buying or transacting in bitcoin by mandate, because of bitcoin’s high relative volatility and the perception by many of bitcoin as a Ponzi scheme. An ETF would NOT address these issues, at least not in the short term. However, it would progressively address both in the long run — by making bitcoin more liquid and thus less volatile, and more widely accepted among institutional circles.

It should be noted that some US institutional players could already be offering long or short exposure to bitcoin via an Exchange-Traded Note (ETN) with futures as underlying, just as for example Barclays offers exposure to VIX with futures as underlying. Yet this is not happening. Why? It is my opinion that perceived demand/profits are not large enough to secure mandates. For detailed information on ETNs, see here and here, as well this SEC link covering the differences between ETFs and ETNs.

The “CBOE bitcoin ETF”

The so-called “CBOE bitcoin ETF” is a physically backed bitcoin ETF proposed by the CBOE exchange (physically backed = when investors buy shares, the fund buys tokens). It has been all the rage since late June, fueling a 40% price increase from the year lows (even though in the last week it retraced 60%). It requires the approval of the SEC. And the SEC has a long history of rejecting bitcoin ETF proposals, with the “Winklevoss bitcoin ETF” standing out as the most famous one.

Many believe the SEC will approve the “CBOE bitcoin ETF” because the CBOE is a very serious institution. Even though true, that conclusion is based on a false premise, because there actually is no such thing as the “CBOE bitcoin ETF”.

To be accurate, the CBOE BZX Exchange made a proposal to the SEC to list SolidX Bitcoin Shares issued by the VanEck SolidX Bitcoin Trust — just as in 2016/7 it was the Bats BZX Exchange proposal to list Shares of the Winklevoss Bitcoin Trust. To compare apples with apples, we should be talking about the CBOE BZX proposal vs. the Bats BZX proposal, or the VanEck ETF vs. the Winklevoss ETF.

Furthermore, in March 2017 the CBOE acquired Bats. Hence, presumably the people behind the CBOE proposal are about the same as those behind the Winklevoss ETF. It should become clear that the CBOE proposal is not more reputable because of the CBOE name. Finally, the listing exchange is the same: the BZX Exchange for US equities.

I will from now on refer to the “CBOE bitcoin ETF” as the “CBOE proposal”.

CBOE proposal — involved parties

  • Fund: SolidX Bitcoin Shares.
  • Filer/Exchange: CBOE BZX Exchange.
  • Trust/Fund Issuer/BTC Custodian: VanEck SolidX Bitcoin Trust.
  • Trust’s Sponsor/Manager: SolidX Management.
  • Trust’s Administrator & Cash Custodian: BNY Mellon.
  • Marketing Agent: Foreside Fund Services.
  • Marketing: Van Eck Securities Corp.

ETF mechanics

  • The Trust issues baskets of ETF shares to Authorized Participants, upon Authorized Participants placing an order to buy. Each basket consists of 5 shares.
  • Authorized Participants are registered broker-dealers and FINRA members, and/or participants in Depository Trust Company (DTC — a trust company which performs the functions of a Central Securities Depository).
  • Authorized Participants deliver cash payment to the Trust’s Cash Custodian.
  • Authorized Participants sell shares on CBOE exchange.
  • Trust purchases bitcoin from bitcoin counterparties (OTC and exchanges, both).
  • (the inverse process applies to selling & redemptions)
  • The MVBTCO index represents a bitcoin price derived from constituent bitcoin OTC platforms that have entered into an agreement to provide such information. This index is used to calculate the Trust’s Net Asset Value (NAV).

Trust’s bitcoin counterparties

The Trust intends to trade bitcoin in the OTC market, with the OTC platforms that comprise the MVBTCO Index. The Sponsor expects that it will be more cost efficient to trade in the OTC market rather than on a bitcoin exchange. The Trust therefore expects to conduct most of its trading OTC, primarily on the OTC platforms that comprise the MVBTCO Index. These are all U.S.-based entities.

The Trust would trade as well with the following cryptocurrency exchanges: Bitstamp (Slovenia), GDAX/Coinbase (California), Gemini (NY), itBit (NY), bitFlyer (NY), and Kraken (California).

All counterparties comply with state and/or U.S. AML &KYC regulatory requirements.

The CBOE proposal indicates that bitcoin counterparties include entities that are regulated by the SEC and FINRA as registered broker-dealers and affiliates of broker-dealers. As of right now the only FINRA member seems to be Genesis Global Trading, an OTC trading entity. Coinbase would soon be a SEC and FINRA registered broker-dealer, although it is not yet one.

SEC concerns regarding bitcoin ETFs

In 2017 the SEC rejected two bitcoin ETFs, while in early 2018 multiple firms withdrew ETF filings, upon the SEC questioning whether the proposed ETFs could comply with various rules. One of the two ETFs rejected in 2017 was the famous Winklevoss ETF; its rejection was appealed, and just recently the SEC rejected it again.

The Winklevoss ETF rejection

Bitcoin pioneers Winklevoss Twins — Caricature — Copyright by DonkeyHotey.

The SEC rejected the Winklevoss ETF on grounds that the proposal was inconsistent with Section 6(b)(5) of the Exchange Act, requiring rules of a national securities exchange designed to prevent fraudulent and manipulative practices and to protect investors.

The SEC argued that to be consistent with Section 6(b)(5), given that bitcoin markets are not resistant to manipulation, the listing exchange must enter into surveillance-sharing agreements with a significant bitcoin market, which must be regulated.

The Commission agrees that, if BZX had demonstrated that bitcoin and bitcoin markets are inherently resistant to fraud and manipulation, comprehensive surveillance-sharing agreements with significant, regulated markets would not be required, as the function of such agreements is to detect and deter fraud and manipulation. But because the underlying commodities market for this proposed commodity-trust ETP is not demonstrably resistant to manipulation, BZX, as the ETP listing exchange, must enter into surveillance-sharing agreements with, or hold Intermarket Surveillance Group membership in common with, at least one significant, regulated market relating to bitcoin.

(An ETP is an Exchange-Traded Product; for the purposes of this article the term ETP is interchangeable with ETF)

Finally, the SEC stated that any surveillance-sharing agreements are with bitcoin-related markets that are either not significant, not regulated, or both.

Finding that BZX has not demonstrated that bitcoin and bitcoin markets are inherently resistant to manipulation, the Commission subjects the proposal to the analysis it has historically used to analyze commodity-trust ETPs, focusing particularly on whether there are comprehensive surveillance-sharing agreements with significant, regulated markets. Because adequate surveillance-sharing agreements are not in place — and any current surveillance-sharing agreements are with bitcoin-related markets that are either not significant, not regulated, or both — the Commission concludes that the proposal is inconsistent with Exchange Act Section 6(b)(5).

The following sentence, from the initial rejection of March 2017, is notable:

The Commission believes that the significant markets for bitcoin are unregulated and that, therefore, the Exchange has not entered into, and would currently be unable to enter into, the type of surveillance-sharing agreement that helps address concerns about the potential for fraudulent or manipulative acts and practices in the market for the Shares.

Comparison with a gold ETF

In its proposal BZX contended that the SEC approved the streetTRACKS Gold Shares ETP even though the spot gold market were largely unregulated. Yet with the Gold order the SEC determined that “the unique liquidity and depth of the gold market, together with the Memorandum of Understanding with NYMEX/COMEX (gold futures market) and NYSE, created the basis for the ETP listing exchange to monitor for fraudulent and manipulative practices in the trading of the Shares”. Bitcoin markets on the other hand are, at present, considerably illiquid and accordingly volatile.

Other concerns the SEC wants addressed

In January 2018 the SEC sent a Staff Letter on Fund Innovation and Cryptocurrency-related Holdings, identifying questions it would want satisfactorily addressed by ETF proposals.

  • Valuation: ETFs must value their assets each business day to strike a net asset value (“NAV”). Can this be achieved given cryptocurrencies’ volatility, fragmentation and lack of regulation of underlying markets?
  • Liquidity: ETFs must maintain sufficiently liquid assets to provide daily redemptions. What steps would funds take to ensure they would have sufficiently liquid assets to meet redemptions?
  • Custody: Funds must maintain custody of holdings. How would a fund satisfy custody requirements? How would a fund validate existence, exclusive ownership and software functionality of private keys? To what extent would cybersecurity threats impact safekeeping of assets?
  • Arbitrage: An ETF is required to have a market price that would not deviate materially from its NAV. In light of the fragmentation, volatility and trading volume of the cryptocurrency marketplace, how would ETFs comply? How would volatility-based trading halts on a futures market impact arbitrage? How would the shutdown of a cryptocurrency exchange affect it?
  • Manipulation: Cryptocurrency markets feature substantially less investor protection than securities markets, with greater opportunities for fraud & manipulation. Is the ETF appropriate for the wide range of investors, including retail investors? Would investors have sufficient information to understand the risks?

Summary of SEC concerns

Primary concerns:

  • Protecting investors, retail investors in particular.
  • Manipulation in underlying markets.
  • Unregulated underlying markets.

Secondary concerns:

  • Daily valuations / NAV.
  • Liquidity availability for redemptions.
  • Custody.
  • Effective arbitrage.
  • Impact of volatility & fragmentation on the preceding issues.

Advantages & shortcomings of CBOE proposal

The CBOE proposal is vastly superior to prior ETF proposals — it is more detailed and addresses most of the issues the SEC has expressed regarding bitcoin ETFs.

Issues addressed

  • Protecting retail investors: by having its per-share price equivalent to 25 bitcoin, the ETF would be cost-prohibitive for smaller retail investors — note this reduces the ETF’s bullish impact as it reduces its addressable market size.
  • Insurance: the ETF would have $25 million in primary coverage and $100 million in excess coverage, with the ability to increase coverage depending on the value of bitcoin held by the Trust, with the goal of maintaining coverage at a one-to-one ratio. On the other hand, the Winklevoss ETF had no insurance coverage “due to insurers’ lack of understanding and sophistication with respect to Digital Assets”.

Issues partially addressed

  • Valuation: the Trust’s NAV would be calculated using the MVBTCO Index, which relies on prices from several OTC market participants. This should be seen as a superior approach (less prone to manipulation) than relying solely on Gemini Exchange prices, as the Winklevoss ETF planned to do.
  • Custody: although not addressed directly in the CBOE proposal, custody solutions are nowadays superior than in 2017.

Issue inconclusively addressed: Manipulation

The CBOE proposal attempts to convince the SEC that manipulation is a non-issue, by means of sentences such as “bitcoin is not particularly susceptible to manipulation” and “it is generally not possible to disseminate false or misleading information about bitcoin in order to manipulate”.

That could not be further from the truth. For example, wash trading, a form of market manipulation in which a market participant simultaneously sells and buys an asset to create misleading activity in the marketplace, is believed to be rampant across crypto exchanges.

Furthermore, false or misleading information about bitcoin is often disseminated. Such events do move markets, as was the case with the bearish misleading news about Korean regulations last January, and the bullish misleading news about Blackrock planning to enter the crypto market.

Issue inconclusively addressed: Unregulated underlying markets

The ETF’s chosen counterparties are registered with FinCEN (Financial Crimes Enforcement Network) as Money Services Business (MSB), and comply with AML & KYC regulatory requirements. However, beyond AML & KYC, counterparties are mostly unregulated (note that the BitLicense is issued by the New York State Department of Financial Services (NYSDFS), and it is a business license mostly centered around AML).

This is what the SEC said about the BitLicense in the Winklevoss ETF rejection:

Although BZX asserts that it has entered into a comprehensive surveillance-sharing agreement with the Gemini Exchange … the Gemini Exchange is supervised by the NYSDFS, the record does not establish that the Gemini Exchange is a “regulated market” comparable to a national securities exchange or to the futures exchanges that are associated with the underlying assets of the commodity-trust ETPs approved to date.

This is meaningful. For example, the SEC has near real-time access to trade data from US national exchanges, which are exchanges registered with the SEC. Similarly, US futures exchanges provide the CFTC with extensive trade data and clients’ data (see this and this).

Meanwhile, the CFTC had to subpoena crypto exchanges for trade data, as exchanges refused to provide such data to assist a probe into market manipulation. The exchanges that refused to provide the requested information are Coinbase, Bitstamp, itBit and Kraken, whose prices are used for the CME CF Bitcoin Reference Rate, which is in turn used for the CME bitcoin futures expiry. These are the same exchanges the CBOE proposal mentions as partners.

Contrast the ETF proposed by the CBOE to a “regular” ETF. In the latter, Authorized Participants deliver the underlying (stocks, gold, etc.) to a Custodian, rather than cash. Also, Trusts buy the underlying strictly from Authorized Participants who, once again, are mostly registered broker-dealers, such as JP Morgan & Deutsche Bank — companies so heavily monitored that they have their phone lines recorded and chats policed in REAL-TIME.

Finally, consider that even though bitcoin and ethereum are not considered securities, the SEC considers most tokens as securities, and thus most crypto exchanges are non-compliant with the SEC. The SEC has specifically said that exchanges that trade securities, including cryptocurrencies that are securities, must register with the SEC as a national securities exchange. The ETF counts Kraken among its potential counterparties — Kraken trades altcoins, many of which would be categorized as securities by the SEC, and is thus non-compliant with the SEC.

Bitcoin Trading Volumes Analysis: Regulated vs Unregulated

— this section will be added later.

Summary

First, bitcoin markets are still not demonstrably resistant to manipulation. Second, although the Trust does have surveillance-sharing agreements with regulated markets for trading the underlying commodity, these markets are not of significant size. These markets are also mostly unregulated beyond complying with AML & KYC requirements (and are thus in a position to refuse data sharing). Third, the significant markets for bitcoin are currently still unregulated.

Therefore, even though the CBOE ETF proposal is vastly superior than prior proposals, this should likely not suffice.

Comparison with CME & CBOE bitcoin futures

Last year two bitcoin futures contracts were listed on CME and CBOE, both CFTC registered exchanges. Both CME and CBOE self-certified that the contract complies with the Commodity Exchange Act (CEA). Self-certifications satisfied the CFTC’s threshold concerns and got the CFTC’s approval. CFTC rules prohibit exchanges from listing contracts susceptible to manipulation and require exchanges to monitor market activity to detect and prevent manipulation in the cash settlement process.

Some may thus argue that the CFTC has determined bitcoin markets are not susceptible to manipulation. Yet the CFTC has stated that the self-certification for bitcoin futures “does NOT provide for … value judgments about the underlying spot market”.

On the rejection of the Winklevoss ETF appeal, the SEC stated “it cannot conclude that actions taken to date by the CFTC determine whether the proposed bitcoin ETP is consistent with the applicable requirements of the
Exchange Act”. This stance is unlikely to change.

An SEC commissioner supported the Winklevoss ETF

Heston Peirce, one of the four SEC commissioners, expressed dissent with the decision to reject the Winklevoss ETF (Bats BZX proposal). One can expect her to support the CBOE proposal as well. Her arguments were the following:

The SEC’s rejection is based on the proposed ETF being “inconsistent with Section 6(b)(5) of the Act, which requires, in part, that the rules of a national securities exchange be “designed to prevent fraudulent and manipulative acts and practices to protect investors and the public interest.” The Commission focuses its decision not on the ETP shares to be listed on the exchange but on the underlying bitcoin spot market…The Commission erroneously reads the requirements of Section 6(b)(5). The disapproval order focuses on the characteristics of the spot market for bitcoin, rather than on the ability of BZX — pursuant to its own rules — to survey trading of and to deter manipulation in the ETP shares listed and traded on BZX. Section 6(b)(5), however… says nothing about looking at underlying markets, as the Commission often has done in its orders.

Let’s then have a look at Section 6(b)(5) of the Exchange Act:

The rules of the exchange are designed to prevent fraudulent and manipulative acts and practices, to promote just and equitable principles of trade, to foster cooperation and coordination with persons engaged in regulating, clearing, settling, processing information with respect to, and facilitating transactions in securities, to remove impediments to and perfect the mechanism of a free and open market and a national market system, and, in general, to protect investors and the public interest; and are not designed to permit unfair discrimination between customers, issuers, brokers, or dealers, or to regulate by virtue of any authority conferred by this title matters not related to the purposes of this title or the administration of the exchange.

Commissioner Peirce seems to be right. Section 6(b)(5) does not relate to an ETF’s underlying markets, but to the exchange where the ETF is listed. Yet looking at underlying markets is how the SEC has been historically operating. Would the Commission change its approach? Could it be argued that by listing an ETF where underlying markets are prone to manipulation the exchange would be exposing ETF investors to manipulation and thus failing to “prevent fraudulent and manipulative acts and practices” and failing “to promote just and equitable principles of trade”?

It’s important to note that the BZX proposal (as well as other bitcoin ETF proposals) was similar to prior commodities ETF proposals. Therefore the SEC evaluated the BZX proposal using the same standards applied to prior commodities ETF proposals. And in prior proposals, the potential for manipulation of the underlying market was a crucial factor for the decision making process.

Heston Peirce’s second reason for dissent was that the SEC’s “approach undermines investor protection by precluding greater institutionalization of the bitcoin market. More institutional participation would ameliorate many of the Commission’s concerns with the bitcoin market that underlie its disapproval order”. This is partially true.

In line with this line of thinking, the BZX proposal made a comparison with copper, stating that when SEC approved the iShares Copper Trust, “the Commission found that demand from new investors would broaden the investor base in copper and thereby reduce the risk of collusion among copper market participants”.

However, the SEC stated in response that “the Copper Order specifically noted the existence of surveillance-sharing agreements not only between the ETP listing market and copper futures markets, but also between the ETP listing market and a significant copper spot market, the London Metal Exchange”.

Given this precedent, it would would be rare for other commissioners to ignore their reservations about manipulable unregulated underlying markets and bypass Exchange Act Section 6(b)(5), in order to promote institutionalization that would ameliorate such reservations.

True, there would be no faster way to make bitcoin markets less vulnerable to manipulation than opening the market to institutional investors’ liquidity. However, faster is not necessarily better, as “faster” may not comply with pre-existing rules. Bitcoin markets could also be made less prone to manipulation by having most of its constituent exchanges regulated. Institutionalizing bitcoin by approving an ETF is not required for diminishing the effectiveness of manipulative practices.

Timing the SEC announcement

The SEC is required to announce a decision on the CBOE ETF proposal WITHIN 45 days of the date the proposal notice is published in the Federal Register. The decision can be postponed by up to 90 days from filing, then up to 180 days from filing, and then up to 240 days from filing. Such timelines are applicable to all filings with the SEC. The ETF’s notice was published on Jul/2/18, thus the first deadline is Aug/16/18, and the final deadline is Feb/27/19.

UPDATE: On Aug/7/18 the SEC announced it is postponing the decision by 45 days, moving the deadline from Aug/16/18 to Sep/30/18. This announcement should have been fully priced-in and represent no news, yet bitcoin markets are so inefficient that the market tanked 6% in 90 minutes on the news.

To illustrate how much of a non-event this should have been, look at the wording of the SEC’s Aug/7/18 communique postponing the CBOE BZX bitcoin ETF proposal by 45 days (here), and contrast if with the wording of the SEC’s Aug/23/16 communique postponing the Bats BZX bitcoin ETF proposal by 45 days (here). Aside of a few edits, replacing a names and dates (e.g. CBOE replaced Bats, VanEck replaced Winklevoss) the wording is IDENTICAL.

Trading the SEC announcement

On Mar/10/17 the SEC rejected the Winklevoss ETF seconds before 16:03 EST. Bitcoin’s price dropped 22% in the following 3 minutes (Coinbase data).

March 10, 2017: $BTC drops 22% on the back of the SEC’s Winklevoss rejection.

One should expect a large move and continuation once the SEC approves or rejects the CBOE proposal. Yet required reaction time may be as short as one second.

The fastest way to learn about such news is having a Bloomberg terminal. For those without one, Twitter is great. Back in Mar/10/17, the BBG “SEC REJECTS BITCOIN ETF” headline started hitting Twitter at 16:04 EST.

Should be noted that the 2017 rejection took place in the midst of a strong bull market, and bitcoin prices quickly recovered. I’d not expect such recovery if the SEC rejects the CBOE ETF. On the other hand, if the ETF is approved, I’d expect strong and uninterrupted continuation.

March 10, 2017: $BTC soon continued with its upward trajectory.

Conclusion

As described, the SEC’s concerns regarding market manipulation and unregulated bitcoin underlying markets remain. That aside, if/when approved, a bitcoin ETF would represent a very strong bullish catalyst. Markets are discounting mechanisms. Large players with a bullish outlook need to pre-position themselves, as bitcoin’s low liquidity would make it very difficult for a large market participant to enter the bitcoin market at a favorable price once the ETF is approved — if so one should expect a very strong immediate push up.

Will the SEC follow its rules and abide strictly by prior practices and by the law as stipulated in Securities Exchange Act of 1934? Helping people make money is not among the SEC’s mandates, is it?

Thanks to Jonathan Cheesman for his thoughtful feedback.

Before you go…

If you enjoyed reading this, please consider showing your support by clicking on the Clapping Hands button — the more the merrier, it increases visibility. Can share the article via the following links: Facebook | Twitter | Reddit | LinkedIn | Whatsapp | Email. You can also follow me on Twitter to stay connected. Thank you.


The Bitcoin ETF: breaking down the CBOE ETF proposal was originally published in Hacker Noon on Medium, where people are continuing the conversation by highlighting and responding to this story.

The Age of Economic Anomaly

Forex stock trading

Crypto Disrupted Episode 30: An interview with economist Andrei Polgar, author of http://ageofanomaly.com

An interview with economist Andrei Polgar, author of http://ageofanomaly.com. In the interview Trent and Andrei discuss the state of the world economy, and cryptocurrency.

Also available on iTunes.

For more episodes of Crypto Disrupted subscribe on YouTube, or subscribe and listen on iTunes.


The Age of Economic Anomaly was originally published in Hacker Noon on Medium, where people are continuing the conversation by highlighting and responding to this story.

How to build an accurate Visual Search

Forex stock trading

One of the most popular phrases today is “Google it”. Searching on the web is a spontaneous, unplanned activity for all of us. We instinctively hunt on search engines knowing that in a click of a button we’ll get an overwhelming response in milliseconds.

But how many of us have ever stopped to think how Google gives us exactly what we’re looking for? Google has perfected the search and ranking process by making continuous iterations and tweaks to their algorithm. They have major updates once or twice a year which significantly impacts the search outcome.

“Possum” is one such major update by Google. In 2016, Google ensured that results were dependent on the searcher’s location i.e. your results will be close to your location. So, if you search for a pharmacist, the one closest to you will pop up as the first result. Prior to this update the search result would probably be a list of pharmacists in your area with the 1st ranked store being that which was most searched for. This localisation of search results made it much easier to find places close by without having to specify the area, etc.

Besides these major updates, we are all aware of the fact that the more you search, the overall quality of results improves. The quality of search itself is a spectrum i.e there are no good or bad search results. The more you search on one topic, the results move from “good” to “better” to “best”. In other words, search results only improve over time.

Similarly, visual search is no different than normal text search. The results are bettered with each upgrade of the visual search engine; although what hinders visual search results are not similar to text search.

Let’s look at some of the challenges we have overcome so far –

Depth of Subcategories

Let’s say we search for jeans as seen in the picture below. There are a number of different kinds like flared, boyfriend, distressed etc. that the visual search engine may pick up. But our search engine picks up jeans only like the ones searched for. The subcategories of jeans is a challenge which our visual search engine has overcome. They train the neural network powering the visual search engine with multiple subcategories of articles. Currently, we support more than 1000 subcategories of articles like apparel, accessories, furniture and kitchen items sold online.

Ability of a Visual Search engine to recognise various sub-categories of jeans

Model Posture

This is by far the most common challenge faced by most visual search engines. As seen below, the query is of a suit worn by a model, posing in a certain way. The image search for this can bring back less relevant results with images of models posed in the same way instead of the clothes they are wearing. However, we train our neural network with hundreds of images each with a different pose to make sure it recognises clothes regardless of the pose. Therefore, our results as seen in the picture below is of other suits and not the model’s pose. Older search algorithms like the one based on locality sensitive hash based algorithms cannot achieve this.

Ability of an image search engine to recognise a blazer in various pictures where the model poses differently

Background noise

As the name suggests, if the picture being searched has a vivid background, it is very possible that the search result will only bring images of matching backgrounds instead of the desired object. Below we can see search results with varied background images rather than the patio furniture being searched for. We have trained our neural network to remove backgrounds from the images. This way the search is only performed on the main object.

An example where the search engine provided results with a strong bias on background instead of the object

Multiple item query

Your search could be for furniture for the entire room as shown in the picture. While searching for multiple articles, our search engine lets us select which pieces we want to search for by highlighting them with boxes or by using the auto crop feature i.e. it lets us select the object of interest easily by automatically cropping parts of the photograph not needed. Most of the other existing solutions would have results with entire combinations as depicted in the picture which are often not the most relevant results.

Example where Auto-crop helps Visual Search engine serve better results

Color weightage challenge

Suppose we are searching with an image of a floral patterned, yellow shift dress. Our actual expected result might be a red dress with similar floral patterns but all the visual search engine gets back to us is with more yellow dresses. This challenge is overcome when the visual search engine has color detection decoupled from other parameters. Our results are not be only based on the color, they are based on pattern, sleeve length and other attributes to the dress. Therefore, we ensure that color is not the only parameter used while searching for visually similar products.

Conclusion

For us overcoming these obstacles has become a regular and planned activity. We take 10 diverse images of each subcategory, pass it through our visual search engine and check what the results are. Referring to our previous example, we will pass 10 different types of jeans and test what the results will be. This ensures our search results do not have the above explained biases in any of the subcategories which we support.

Similar to Google, Turing Analytics also has scheduled updates every three months for major changes. For their last one in May 2018 they doubled the number of subcategories they support. Prior to this in Feb 2018 they released their auto crop feature allowing them to overcome the challenge of multiple items in one image. Another important update they had was in December last year where they reduced the search response time by approximately 80%. Turing Analytics’ search engine is a revolutionary software enhancing customer satisfaction.

Now that you know so much about how accurate Visual Search results are at Turing Analytics why not try our demo at VisualSearch.App/Demo


How to build an accurate Visual Search was originally published in Hacker Noon on Medium, where people are continuing the conversation by highlighting and responding to this story.

Paying at the Pump 

On a recent road trip, this writer learned that in Oregon it is “against the law to pump your own gasoline,” as the designated pumper put it. What penalty a motorist would incur for daring to pump her or his own gas was not apparent, but according to the attendant, it was a matter of the government creating jobs. Any motorist could be forgiven for thinking it counterproductive to pay someone to perform a task just about anybody could do. At one time, some gasoline outlets offered a choice of “full service” or “self-serve,” usually at a lower price. In Oregon and New Jersey, the only states to ban self-service, it’s the government’s way or the highway. After the initial shock, out-of-state motorists find this a matter of some amusement, but there are parallels.

As this writer’s grandchildren can verify, even a four-year-old can select a floor and push the button on the elevator. Yet, at one time the California legislature was paying government employees to operate the elevators, which go to the right floor and make a perfect level stop every time. In a better arrangement, government pays workers for tasks it is not practical or prudent for individuals to perform, such as police and military service. When the government pays for something anyone can do it wastes taxpayer dollars and increases prices. As this motorist noted, prices were much lower at self-serve stations in Washington State, under $3.00 at one, and 18.4 cents of that is federal tax alone.

At one time, motorists of a certain age will recall, gasoline outlets were “service stations.” A uniformed attendant would pump the gas, check the oil, check the tire pressure and clean the windshield, all with amazing efficiency. In Oregon and New Jersey, all they do is pump, so their ban on self-service isn’t exactly a progressive policy. 

***

K. Lloyd Billingsley is a Policy Fellow at the Independent Institute and a columnist at The Daily Caller.

copy top traders